Acute stress and long-lasting blood pressure elevation: a possible cause of established hypertension?

Although there is no doubt that stress, in its various forms, can induce transient elevations of blood pressure [1-3], the question of whether an acute stressor of sufficient strength might lead to a persisting blood pressure rise, therefore inducing or aggravating an arterial hypertension condition, is still a matter of debate [1,4-6]. Furthermore, the role played by the specific nature of the stressor that is considered to determine a significant and long-lasting increase in blood pressure has been, and still remains, a matter of lively discussion [1,7,8]. An example of the importance of stress-related mechanisms in developing high blood pressure, and in leading to potentially life-threatening cardiovascular complications, is provided by the haemodynamic effects of a number of real-life stressors, including natural disasters such as earthquakes. Previously, Parati et al. [9] were able to provide direct evidence for this, by recording the direct effects on ambulatory blood pressure and heart rate of a moderate-intensity earthquake that struck central Italy. The effects of the earthquake consisted of a 20% increase in systolic blood pressure, a 46% increase in diastolic blood pressure and a 79% rise in heart rate. This indicates that the cardiovascular response to a natural stressor of this kind involves both cardiac and vascular stimulation, presumably triggered by an increase in sympathetic activity. This sympathetically-mediated combined increase of blood pressure and heart rate, and thus of their product, may represent one of the pathophysiological mechanisms responsible for the increased rate of cardiovascular events (i.e. mainly acute myocardial infarction and sudden death) that are reported to occur during an earthquake [10-14]. Moreover, abnormal blood pressure values persisted for approximately 1 h after the quake and were followed by a long-lasting period of enhanced blood pressure variability and a blunted nocturnal blood pressure fall. This is in line with previous reports indicating that a major daily life stressful event of sufficient intensity (e.g. a medical student attending an important university interview) [3] may markedly and persistently alter ambulatory blood pressure values, leading to a disruption of the usual 24-h blood pressure profile. The resulting increase in blood pressure variability, and the concomitant blunting of nocturnal blood pressure fall, might further contribute to the increase in cardiovascular risk typical of this condition [15-18]. In this issue of the journal, Gerin et al. [19], investigated the potential influence of another major type of real-life stressor (i.e. the attacks in New York City on 11 September 2001) on the magnitude and the time course of blood pressure elevation in patients with known or suspected hypertension. The study investigated subjects who, at the time of the attacks, were participating in a clinical trial on telemonitoring of blood pressure and medication compliance at four sites in the USA, two of them (New York City and Washington DC) close to, and two others (Chicago and rural Mississippi) remote from, the sites of the attacks. Gerin et al. [19] report on the occurrence of variable but significant and sustained increases of systolic (but not diastolic) blood pressure in the 2 months of blood pressure-monitoring after 11 September compared to the same time period before the event. The authors also provide evidence that this observation was not merely due to seasonal changes that are known to result in a blood pressure elevation during the colder periods of the year. Indeed, the stress-induced blood pressure increase was also evident compared to blood pressure values measured during the corresponding time period of the year preceding the attacks. In spite of its interest, there are some flaws in the design of their study, mainly due to the necessarily retrospective approach, with there being a limited number of individual patients from whom blood pressure values were available both before and after September 11. The results therefore rely heavily on comparisons between groups in which blood pressure was monitored either before or after the event only. Curiously, but not entirely surprisingly, the estimated increase of systolic blood pressure, after adjusting for confounding dependencies among the collected observations and for trends in the data, was lowest in New York (1.58 mmHg) and highest in Washington DC (8.67 mmHg). Naively, the greatest rise in blood pressure would have been expected in the city with the highest number of casualties and most eyewitnesses, although this was not the case. However, despite these shortcomings, the findings lend further support to the hypothesis that a powerful real-life stressor is able to induce a persistent increase in blood pressure, outlasting the actual influence of the stressful event on the subjects exposed. Against the background of these observations, two important questions concern, on the one hand, the nature of the stressor responsible for the blood pressure rise in this case and, on the other hand, the mechanisms involved in such a long-lasting blood pressure response to stress. We can define this type of stress as ‘psychological’. In a recent review by Lovallo and Gerin [20], psychological stressors were defined as ‘… events that challenge the homeostasis of the organism because of their perceived threat value, regardless of potential for physical harm’. In their review, the authors describe three levels of response to psychological stress that may contribute to the observed degree of cardiovascular reactivity: (i) the cognitive-emotional level, for which the anatomical basis resides in cortical and limbic structures located above the hypothalamus, which is responsible for individual psychological differences in cognitive-emotional responses; (ii) the autonomic-endocrine level, regulated by anatomical structures localized in the hypothalamus and the brain stem, which amalgamates the input from higher brain centres with endocrine and autonomic responses; and (iii) the peripheral level, including peripheral sources for the different individual reactivity to various inputs. An example of these peripheral factors are the structural changes of arterial walls, leading to altered vascular reactivity in response to external inputs. A great number of experimental results in animals and in humans support the notion that relatively long-lasting blood pressure increases can be evoked by various stressors, acting at any of the three levels described above. Once more, the relevant question is whether these stress-induced increases in blood pressure are likely to be maintained over more prolonged periods of time, resulting in a persisting hypertension condition. An appealing theory, which may provide suitable support for such a possibility, is that proposed several years ago by Björn Folkow, based on extensive animal experiments [21]. In his view, repetitive (stress-induced) sympathetic stimulation, which is able to induce transient rises in blood pressure, heart rate and cardiac output, will lead to structural adaptations of the blood vessels and the heart. The ensuing vascular hypertrophy will engender disproportionate blood pressure reactions to any given stimulus and, in a vicious circle, this will aggravate vascular remodelling, finally leading to permanently elevated vascular resistance and the transformation of the transient increases of blood pressure into permanent hypertension. As with all theories relating stress to sustained hypertension, this hypothesis leaves a number of questions unanswered. For example, why would some individuals react to psychological stressors according to the above-described vicious circle, and develop sustained hypertension according to Folkow's theory, whereas others do not? Are all subjects who develop sustained hypertension characterized by an impairment of autonomic and humoral mechanisms involved in cardiovascular regulation? Or are they mostly characterized, at a higher level in the central nervous system, by an impairment of the cortical and limbic processing of emotional responses, leading to an inappropriate sympathetic outflow? Are these impairments genetically determined, with their appearance requiring a specific genetic background? Is there the possibility of tracing the development of pro-hypertensive mechanisms through the appearance of an alteration in neural transmitters or in transmitter interactions at the brain level, involving serotonin, acetylcholine, opioidergic mechanisms or substance P [22-25]? Finally, do subjects, who develop hypertension under the effects of a major stressful condition, also display alterations in the function of another important blood pressure-regulating organ (i.e. the kidney), as appears to be the case in the spontaneously hypertensive rat [26], which is a widely used animal model of essential hypertension? These questions have been repeatedly asked over the years but, for various reasons, no satisfactory answer has been provided to date. One possible explanation for such a failure is that, because of the high level of complexity in these mechanisms, important methodological problems still need to be appropriately solved. These include the difficulty in distinguishing between emotion and ‘reaction’; the methodological problems in dealing with the different hierarchies of information and transmission in the nervous system; the interplay between genetics and the environment; and, last but not least, with the integration of neuronal, endocrine and autonomic circuits. Another possible reason for the lack a suitable solution to these issues is that, in recent years, a large number of researchers who were formerly active in this field have moved away from the integrated psychosomatic-physiological approaches, which are required to properly answer these questions, and instead have focused mostly on molecular and cellular studies. Cells are certainly characterized by highly attractive complex signalling pathways, and by a genetic apparatus of great scientific appeal but, unfortunately, they do not appear to be able either to express emotions or to develop arterial hypertension. Thus, although the genetic background and molecular mechanisms undoubtedly need to be intensively addressed, they need to be embedded into a framework that integrates them with the complex mechanisms involved in blood pressure regulation, including psychosomatic factors.